How can we go back to the thread where we all have our top ten projections? taking out Rosario and having Bush in that place, i think i fared very well against BA,

This is BA's review of their pre-2003 list. The pre-2004 list comes out on Tuesday.

The 2004 list is out for suscribers and somehow is in a post.
Rios, MacGowan, Quiroz, Gross, Rosario, Hill, Bush,Perkins, Adams, League.

I'm really surprised to see Rosario so high on the list and Arnold dropped. It seems a bit silly to me to rank a pitcher so high after a major injury. A handful of bad starts by Arnold at AAA and he drops off the list? To me, Arnold has at least proven himself at AA, whereas Perkins hasn't mastered high-A yet. BA's probably gives Perkins the nod because he throws harder than Arnold.

This list is indicative of BA's obsession with "high-ceiling" players. To tell you the truth, I've never understood what that's supposed to mean - high-ceiling. I'm sure that Greg Maddux wasn't considered a high-ceiling pitcher when he was in the minors, but he became one of the greatest pitchers ever. My point is that you can't tell what a player's ceiling is, so why not focus more on performance with a nod to tools.

That's what John Sickels does, and I consider his prospect lists to be the best in the business.

While Arnold did struggle in AAA, I don't think he was SO bad that he deserves to be wiped off the Jays Top 10 Prospect list. I agree with the BA list almost unanimously, and while I'm far from an expert, I would think that Arnold has shown much more than a guy recovering from Tommy John surgery who didn't pitch this season. Arnold's ERA in AAA isn't that bad, and while many other peripheral stats look grim, many players do take time to make the adjustment from AA to AAA. I don't think it's time to cast off Arnold just yet...

(I wrote this and posted before i saw what Robert Dudek wrote --- I guess we're both thinking the same thing ;) )

I agree that it's not time to cast Arnold off, but he didn't make my top 10 prospect list (in a deep organization) for the following reasons:

1. his performance in AAA was not just a few bad starts; the decline in his K rate, and the increase in his HR rate were significant,
2. he doesn't have overpowering stuff, and
3. he'll be 25 in May.

I think he's got a chance to put it back together in Syracuse, and then have a shot late in the season in Toronto, but right now, he's about at the same place as Mike Smith was last year.

Incidentally, Robert, I checked back on Greg Maddux. While BA might not have called him a "high ceiling" prospect, I would have. At age 20, he pitched 128 innings at triple A allowing 3 homers and 30 walks, while striking out 65. My major concern about him would have been the large number of innings he threw at age 19 and 20.

All that said, I too like John Sickels' approach which marries the objective and subjective quite nicely.

http://www.gerrymcdonald.ca/index.html
Mike, I'm not sure it's fair to compare Arnold to Smith. While age and performance are comparable at the AAA level, keep in mind that:
a) Smith was never really regarded as a decent prospect while Arnold has been always regarded as having a lot of talent and potential --- hence he was drafted in the 2nd round
b) While Arnold is 25 in May, as a college draftee in his senior year, he hasn't been in the minors for all that long (Bush will be in the same situation and I don't think that diminishes his status all that much.) While I'm not sure about Smith, I assume he's been in the minors far longer.
c) Smith never dominated A and AA competition the way Arnold did.

Also...I entered a random search for Mike Smith and while Baseball Cube doesn't seem to be working right now, I did find another Blue Jays Minor League site. I haven't looked at it much yet, but it seems alright...(COMN for link)

I agree with Mike Green on Arnold though I still would have him ahead of Vince Perkins on my list. I guess it shows the depth of the Jays farm system that we can even have this debate and also the higher rate of success in coming back from Tommy John surgery that Rosario still ranks so high. Baseball America seems to be shifting a bit more towards performance as Zach Greinke dosent have overpowering stuff but is ranked very high based on his control and his performance. Look how low Rios was last year and now after his performance he has risen up to the top. I do like Sickels approach as well but all in all I still BA does a bang up job.

...the site seems to acknoledge Batter's Box, so I guess most of you probably have seen it...

One other note on Maddux in 1985 he was the Cubs 5th rated prospect, 1986 the 7th rated prospect and in 1987 the 5th rated. Some of the pitchers rated ahead of him those year, Drew Hall, Carl Hamilton, Gary Parmeter and John Abrego. All household names now. Baseball Amercia archives every teams top 10 lists though you can only read the comments from the year 1999 and on. Mike Smith has never made the Jays top 10 list so I dont think it fair to compare him to Arnold quite yet.

Rated 5th, 7th and 5th - no I don't think BA would have considered Maddux a very "high-ceiling" or high upside pitcher. Those guys get ranked #1 or #2.

Would the Gerry in GerryMcDonald.ca be the same regular at Battersbox?

Would the Gerry in GerryMcDonald.ca be the same regular at Battersbox?

It sure would, Adam. Robbie Goldberg is the first to mention Gerry's site here, but I know that at least one Jays insider and family or friends of more than one minor-league player have already discovered it. It's officially affiliated with Batter's Box, and Gerry will be a big part of our expanded coverage next season.

The Box redesign is taking a bit longer than hoped (these things always do) in part because of a hard drive failure on my PC, which also accounts for my relative silence here of late. When it is complete, there will be a navigation link to a Farm System page, which will include links to our previous Farm Reports and to Gerry's site. I think he's done great work, though some parts, like the depth chart, are still "under construction" and should not be considered 100% accurate. Bear with us; it will all come together soon.

Robert: This list is indicative of BA's obsession with "high-ceiling" players. To tell you the truth, I've never understood what that's supposed to mean - high-ceiling. I'm sure that Greg Maddux wasn't considered a high-ceiling pitcher when he was in the minors, but he became one of the greatest pitchers ever. My point is that you can't tell what a player's ceiling is, so why not focus more on performance with a nod to tools.

Pick a peak value metric X of your choice. Let x be some high value of that metric. Let P(X) be the probability density function of a given player's peak value measured by the metrc X.

A high-ceiling player is one for whom the integral of P(X) over the interval (x,infinity) exceeds some appropriate probability p.

That's how I've always understood it. A high-ceiling player is one who has a relatively high chance of performing above a certain high level. This isn't to say that a "low-ceiling player" has zero chance of performing above a certain level, just that his probability of doing so is lower. The low-ceiling player may still have a higher expected value.

Robbie: I'm not sure it's fair to compare Arnold to Smith. While age and performance are comparable at the AAA level, keep in mind that:
a) Smith was never really regarded as a decent prospect while Arnold has been always regarded as having a lot of talent and potential --- hence he was drafted in the 2nd round
b) While Arnold is 25 in May, as a college draftee in his senior year, he hasn't been in the minors for all that long (Bush will be in the same situation and I don't think that diminishes his status all that much.) While I'm not sure about Smith, I assume he's been in the minors far longer.
c) Smith never dominated A and AA competition the way Arnold did.

(a) is true, but if you use it as an argument, you're acknowledging an important role for "non-objective" scouting in player evaluation. (I know that you don't deny such a role entirely; I'm arguing for a larger one.) And you have to give props to BA for not ranking Smith very high in 2002 (he was 13th) despite his amazing performance (dominating two levels and skipping a third.)
(b) Smith was drafted in 2000 at the age of 22.9. He was probably a senior. He made the majors after less than two full years in the minor leagues.
(c) Smith skipped low-A, and Arnold skipped high-A. Here are their career AA stats (bear in mind that Arnold pitched for a while in the Texas League.):

Pitcher    IP  H/IP HR/IP BB/IP  K/IP
Smith    51.0  0.80  0.02  0.33  1.08
Arnold  110.2  0.70  0.05  0.36  0.94

I don't have stats on batters faced for previous years, so I can't do rates per PA.

Smith was seven months older than Arnold at each step up the ladder, which is a minor point against him. On the whole, I'd say that "objective" evidence would put Smith a little bit behind Arnold every step of the way, but I think he's a pretty good comp.

Jabonoso: Here is the link to the page where we posted our top-10 lists. Here are the results that Robert compiled.

http://tsf.waymoresports.thestar.com/thestar/baseball/player.cgi?2995
John and Robbie, COMN for Mike Smith's career record. His A ball numbers were as good as his AA numbers.

John,

I would agree with your definition of "high-ceiling", but I'm almost certain that BA doesn't use any sort of metric to come up with a quantifiable probability of a given player reaching peak value X. If they did, they'd have to use a player's age and past performance focusing on core skills (such as the ability to strike batters out). That's basically what I do with my prospect rating.

I think BA assesses the tools - grades the fastball and off-speed pitches and comes up with something like a "stuff" rating. It's the guys who score high on the "stuff" scale that are considered "high-ceiling" even those who have severe control problems such that their chances of becoming great pitchers are miniscule.

Perkins has better stuff than Arnold, but Arnold's core skills are more advanced for age than Perkins's. Objectively, Arnold has a higher probability of reaching a high peak, but because his "stuff" isn't as good as Perkins', BA doesn't consider him as good a prospect.

Oops ... my comment should read "Smith skipped High-A, and Arnold skipped Low-A," not vice versa.

Robert,

I would be extremely surprised if BA used a metric in the manner I described. My point is that differentiating between "high-ceiling" and "low-ceiling" (I'm dropping the scare quotes from here on) players is neither empty (in the sense that there's no real difference) nor useless (in the sense that high-ceiling players aren't more valuable in any tangible way.)

I suspect that BA employs performance-based analysis a little bit more than you think, but their emphasis is definitely on traditional scouting. Considering that they have consistently had higher opinions of Rios and McGowan than other analysts, I think that the scouting approach does bring something to the table. I agree with you that Sickels' approach is probably the most balanced; nevertheless, I still think BA's list is quite valuable, even if it might be improved by regression towards a performance-based mean.

I think that it's appropriate to use the phrase "high-ceiling" to describe a guy like Vince Perkins. The chance that he puts it all together and develops real command is probably not all that high, but it's nonzero, and if his stuff is as good as is reported then he could really be something special. Even if he only develops marginal command (which may not be particularly likely), he could be Kelvim Escobar. At the same time, the odds are that he doesn't have any major league career to speak of. By contrast, Arnold will more likely than not have a major league career of some sort, but quite likely as a middle reliever. I would not give him much chance of becoming an above-average major-league starter at this point.

I don't think that we really have the tools to make an evidence-based prediction of who is a better prospect at this point. There are so many variables (age, level, this year's performance, previous years' performance, scouting data, etc.) that in the end we have to use our own judgement one way or another. If we had a validated instrument for predicting future performance (ideally, one that produces a probabilistic prediction), we could defer to it, but I don't know of any such instrument. (I'm unaware to what degree PECOTA has been validated for minor-league pitchers.)

John

John,

It may seem like Perkins has a better chance to be a great pitcher, but it is far from obvious. Other than having better "stuff" what makes you think he has a higher ceiling than Arnold? Plenty of guys who throw 88 to 90 become great pitchers; some pitchers add mileage to their fastball as they reach 27 or 28 (that could be Arnold). Lot's of guys who throw mid-90s at 21 are throwing 88 to 90 when they are 27.

I just don't see what practical use the term high-ceiling could have. If you want to talk about tools that's fine: Perkins has better stuff, Arnold has better command. And Arnold has proven more at a higher level - which is very important WRT pitchers. That makes him a better overall prospect.

I checked back to our forecasts of BA's top ten list. Most forecasters had either 7 or 8 names correct. I did not see anyone with more than 8. Rosario was missed by most, while Arnold was the biggest name omitted from BA's list.

By my reckoning the most accurate forecast was from Kristian. He voted three times so he should have been accurate (actually all three were the same vote).

Here is his list, with the BA rank after his choice:

1. Rios - 1
2. Magowan - 2
3. Quiroz - 3
4. Bush - 7
5. Gross - 4
6. Hill - 6
7. Arnold - nr
8. Perkins - 8
9. Adams - 9
10. Peterson - nr

Six votes right on the mark, two slightly off and two misses.

Well done Kristian

I was looking through the history again of some American League teams and one guy who has always stuck out in my mind as a guy with supposedly average stuff was Mark Buerhle. He was ranked 10th by BA in 2000 behind Kip Wells,Jon Garland, Aaron Myette, Jason Strumm, Matt Ginter, Dan Wright and Lorenco Barcelo. Buerhle was ranked lower despite comments that he could be the fastest moving pitcher because of a fringe fastball.So, BA I think still puts more emphasis on guys who could be dominant pitchers yet BA dosent place as much emphasis on guys who might be more likely to make the majors yet never become Ace guys. Perfect case in point Arnold vs Perkins.

Let me point out that Buerhle HAS become an ace (so has Maddux and Moyer).

If a pitcher is at all successful in the minor leagues then he has a chance to be an "ace". That's why there is no actual "ceiling" in the sense that we can most definitely say that Pitcher X has zero chance of becoming a great pitcher.

I agree with you Robert. If a pitcher in the minors has success he has a chance to be an Ace. Obviously scouts use the tool scale to hopefully enhance their predictions on who has the greater chance to become that ace while stats can be analyzed to draw similar conclusions. How many time has a pitcher dominated Triple A yet he gets hit hard at the majors and becomes a "4A" player. With prospects there is just no fool proof way of knowing for sure thats why depth is so important in an organization. For every Buerhle there are 3 Ginters and so forth. When I evaluate prospects I take age, stats, level of competition and then focus on key stats an example being K to BB ratio for a pitcher or even WHIP. If you look at Buerhle he was dominant in those areas before reaching the big leagues. I also use the scouting reports and tools but any pitcher can throw 98 but if cant throw strikes its worthless. Now that I am a frequent reader of this site it has opened my eyes to a whole new lever of statistical analysis. Buerhle was my example because in a Fantasy I drafted him very cheap because I loved his stats in the minors. Jon Stephens is another guy I would like to see get a chance. Colt Griffin would be the opposite so far as a player with the tools but lack of stats to show for it. That being said BA tends to write off some of these guys too early or at least lower their rankings, I am interested to see how they rate Jeremy Guthrie whose performance was a lot worse than Arnolds at the Triple A level. Robert your earlier formula to rank the pitchers I think showed a lot of both accuracy in comparison to other lists and also showed the success of the some the guys from the past using the formula (Johan Santana I believe was one guy that stuck out). When can we expect a list for the hitters?

Interestingly, on the Jason Arnold debate, Josh Boyd (in the BA chat on the Yankees system) would rank Arnold #8 in the Yankee's system. He also says Toronto's system is stronger than New York's.

I agree with BA's list regarding Arnold 110%. He has been hyped up since coming out of UCF (as are most players coming out of bigger schools).I want to take nothing away from him. He is an exceptional pitcher considering the stuff he has. When comparing his Stuff to that of Perkins or League (who are both ranked higher without the performance results) or even D.J. Hanson for that matter, there really isn't much of a comparison. One more thing to ponder for you guys, If you take away the walks category, I would say Perkins had as good a season as anyone. He led the organization in ERA and was 14 - 7 with a .179 OBA. When he does fix his control issues, which the Blue Jays are sure he will, He is going to be scary.

Well, this is from the Dodgers chat and will give us a good idea of where our system stacks up, at least in their eyes (sorry if this has been reported elsewhere):

Andy from San Francisco asks:
Josh, the Dodgers farm system seems to be flourishing after some down time in the late '90's - How would you compare the LA system w some of the other better farm systems around - What in your opinion are the 5 best systems in the baseball today ? thanks

A: Josh Boyd: Not in order, the Dodgers, Indians, Angels, Blue Jays, Brewers are the strongest.

When he (Perkins) does fix his control issues, which the Blue Jays are sure he will, He is going to be scary.

Where does it say 'they' believe he will? Hopefully he will, absolutely. But it seems pretty presumptuous to say he will, lots of guys don't.

A lot of guys do. I don't call that presumptuous, I call it Positive. Maybe even wishful thinking but why come on here and focus on the negative, right? Yes, eventually he will have to fix the walk totals, lets hope for the benefit of the Blue Jays that he does....

As long as we're being positive, why can't we presume that Arnold will add 3-4 MPH on his fastball?

If Arnold gained velocity and Perkins gained some control, that would be great for everyone. Maybe they can compare notes.

I dont really think Arnold needs to gain velocity as much as he needs experience. He has done very well at each level until reaching Triple A last year. Lets see how he does this year but considering all of his past success I think he will adjust to triple A and at the mid point of this year will be on the Jays roster. A WHIP of 1.39 with a K to BB ration of just under 2 to 1 isnt bad for his first time at the Triple A level.

Actually, if Arnold were to add a few mph's he wouldn't be "gaining", he would be "regaining". In college and during his first stint in the minors Jason Arnold threw as high as 97 mph, remember. As everyone knows, he came out of college as a 'closer' and has started ever since. He could be heard during Syracuse radio broadcasts last year explaining that he dropped some mph so he could become less a 'thrower' and more a pitcher, yada, yada & yada.

Hello Robert Dudek;

I was wondering if you were going to share "The best minor league pitching prospects " this year.

As can be found in the archive at http://www.battersbox.ca/archives/00001235.shtml

You had also mentioned a hitting prospect method have you shared this?